Get Started
Providers

Render Provider

Deploying backend services and databases to Render.

Render Deployment Provider#

Atlas supports triggering deployments for web services and static sites hosted on Render.

Pre-requisites#

Atlas can deploy your project to Render either by linking to an existing service or by automatically creating a new service for you directly from the CLI.

Configuration#

To deploy to Render, you need two pieces of configuration:

1. Authentication#

You need to provide your personal Render API key. You can do this in two ways:

  • Environment Variable: Set the RENDER_TOKEN environment variable in your .env file or shell profile. This is the recommended approach for CI/CD environments.
  • Stored Credentials: Run the interactive CLI to securely store your token locally:

2. Service Creation & Configuration#

You do NOT need to manually provide a Service ID or create the service on the dashboard. When you deploy a project for the first time, Atlas will:

  1. Automatically detect if your project is a Static Site (e.g. React/Vite) or a Web Service (e.g. Node/Express).
  2. Determine your build and start commands by analyzing your package.json.
  3. Use the Render API to automatically create the service on your account.
  4. Save the newly generated render_service_id safely in your local .atlas/project.json configuration so future deploys reuse the same service.

Usage#

To trigger a deployment, use:

Git Push Preconditions#

Because Render builds and deploys your code directly from the connected remote Git repository (e.g., GitHub, GitLab), Atlas enforces a strict push-check for Render deployments.

Before triggering a deployment, Atlas will check that:

  1. Your working directory is clean.
  2. The current commit (HEAD) on your local machine exactly matches the commit on the remote tracking branch (e.g., origin/main).

If you have uncommitted changes or unpushed commits, the deployment will fail fast, prompting you to push your changes first.

Deployment Orchestration#

When you run a deployment, Atlas performs the following steps:

  1. Validates authentication and the Service ID.
  2. Checks your Git tree to ensure all code is pushed to the remote repository.
  3. Triggers a new deployment using the Render API, pinning the deployment to the exact Git commit SHA that you are currently on.
  4. Polls the Render API to monitor the deployment status in real-time.
  5. Returns the live URL of your service once the deployment succeeds (or reports a failure if the build crashes).
  6. Health Check & Rollback: Post-deployment, Atlas verifies the live URL returns an HTTP 200. If it fails, Atlas can seamlessly revert the service by re-deploying the exact Git Commit SHA of your LastHealthyDeployment using Render's deployment trigger API.